Quantity and quality are increasing but there's room for improvement: A scoping review of physical activity intervention trials

•There are 1779 physical activity trials in the physiotherapy evidence database (PEDro).•The mean (SD) PEDro score was 5.3 (1.5) points out of 10, reflecting ‘fair to poor’ quality.•Journal impact factor is weakly correlated with trial quality (0.21, p < 0.001).•We make five recommendations to improve future trial quality. Observing trends in research publications helps to identify the quantity and quality of research produced, as well as reveal evidence gaps. No comprehensive review of the quality and quantity of physical activity intervention trials has been conducted. We aimed to investigate i) the volume and quality (and changes in these over time) of randomized controlled trials evaluating physical activity interventions, and ii) the association between journal ranking and trial quality. We searched the Physiotherapy Evidence Database (PEDro) for trials investigating physical activity interventions (no restrictions for population, comparison, or language). Descriptive statistics were used to describe the volume and quality of trials. The association between journal ranking (Journal Impact Factor) and trial quality (PEDro Scale) was examined using Spearman's rho correlation. We identified 1779 trials, of which 40% (n = 710) were published between 2016 and 2020. The mean (SD) total PEDro score was 5.3 (1.5) points out of 10, increasing over time from 2.5 (0.7) points in 1975–1980 to 5.6 (1.4) points in 2016–2020. Quality criteria that were least reported included blinding of intervention deliverers (therapists) (n = 3, 0.2%), participants (n = 21, 1.2%), or assessors (n = 541, 31%); concealed allocation to groups (n = 526, 30%); and intention to treat analysis (n = 764, 43%). There was a small correlation between trial quality and Journal Impact Factor (0.21, p < 0.001). A large volume of trials has investigated physical activity interventions. The quality of these trial reports is suboptimal but improving over time. Journal ranking should not be used for selecting high quality trials.